Stock Picking:

Any person interested in the stock market has probably heard of or read about dozens of systems and/or signals used for picking the next "hot" stock. Some of the commonly followed methods include "value" investing, "momentum" investing, and the various "technical" signals used by the Wall Street "elves". We have followed the market for over twenty years, as a hobby, and have always been amused by the observation that, for almost any stock picking method you can think of, there was an equal and opposite method that seemed to have similar (and equally ho-hum) results. It is a fact, often observed but rarely actually absorbed, that very few stock pickers, fund managers, or newsletter writers have been able to "beat the market" over an extended period of time.

With the above skepticism well ingrained, it was with great interest that we read Mark Hulbert's article "A Strong Signal" that appeared in the April 22, 1996 of Forbes magazine. This piece discusses a study undertaken at Rice University where the performance of stocks split 2 for 1 is measured in relation to performance of the market as a whole. The final results of the study are published in the Journal of Financial and Quantitative Analysis and would indicate that there is a measurable difference in the performance, for up to three years, of stocks that have split 2 for 1 as opposed to those that have not.

With the help of computers and today's vast databases, we thought it should not be too difficult to put this theory to the test by creating some hypothetical portfolios out of stocks that have split and comparing them to the market. It soon became clear that this was not as easily accomplished as first thought. Of the 7000+ stocks traded on the NYSE, AMEX or NASDAQ exchanges, over 1000 stocks have split 2 for 1 since 1990, many of them more than once. In any given month, anywhere from 3 to 50+ stocks will split, with the average being about 15 to 20 over the last 6 years. So, clearly, the trick was going to be selecting which of the several stocks that split in any given time period would be the ones that we wanted in a test portfolio. The following, without getting into the details, are the principles that we have relied on for making that selection that, as you will see, have yielded some very good results. First and foremost, it must be remembered that the universe from which one is selecting, the group of split stocks, is already assumed to be performing "ahead of the market". So all one has to do is achieve an average performance within that pre-selected group.

1. We prefer companies that are making money "the old fashioned way". In other words, we looked for companies that have real earnings that are growing at a moderate pace.

2. We prefer companies that pay dividends. We view dividends as a hedge against a falling market and as a signal, in itself, that the company's management recognizes for whom they are working.

3. We prefer companies that have reasonable price/earnings and price/book ratios. This is in keeping with one other well-known stock picking method, the "Value Line Survey's" rating system that has had good results over the years.